DFRobot
Raspberry Pi AI Camera - 12.3MP Sony IMX500
· MPN: SEN0651
This compact AI camera module brings on-board vision inferencing to Raspberry Pi projects using the Sony IMX500 Intelligent Vision Sensor. It combines a 12.3...
This compact AI camera module brings on-board vision inferencing to Raspberry Pi projects using the Sony IMX500 Intelligent Vision Sensor. It combines a 12.3-megapixel CMOS image sensor with integrated AI processing, enabling real-time visual data processing without a separate AI accelerator.
It is suited to applications such as object recognition, facial detection, security systems, smart robotics and environmental monitoring. The wide field of view and high-resolution imaging make it useful anywhere you need image capture and machine-learning output from a small Raspberry Pi-compatible module.
The camera works with the Raspberry Pi software stack, including libcamera and Picamera2, and supports tensor metadata output for more advanced image-processing workflows. It also provides flexible image outputs including Bayer RAW10, YUV/RGB ISP output, ROI and metadata.
The shipping list includes one Raspberry Pi AI Camera.
Features:
- Powerful vision AI processing: Integrated AI processor accelerates machine-learning tasks directly on the camera module.
- No external AI accelerator required: Enables AI-powered applications such as real-time image classification or facial recognition with simplified hardware.
- High-resolution imaging: 12.3-megapixel resolution with a wide field of view for high-quality images and video.
- Raspberry Pi ecosystem integration: Compatible with all Raspberry Pi computers and the Raspberry Pi software stack.
- Library support: Designed for use with libcamera and Picamera2.
- Flexible output and metadata support: Supports Bayer RAW10, YUV/RGB ISP outputs, ROI and metadata.
Specifications:
- Sensor: Sony IMX500 Intelligent Vision Sensor
- Resolution: 12.3 megapixels
- Sensor Size: 7.857 mm (1/2.3” type)
- Pixel Size: 1.55 µm x 1.55 µm
- Image Dimensions: 4056 x 3040 pixels (horizontal/vertical)
- IR Cut Filter: Integrated
- Autofocus System: Manual adjustable focus
- Focus Range: 20 cm to infinity
- Focal Length: 4.74 mm
- Horizontal Field of View: 66° ±3°
- Vertical Field of View: 52.3° ±3°
- F-stop: f/1.79
- Infrared Sensitivity: No
- Image Output: Bayer RAW10
- Image Output: ISP output (YUV/RGB)
- Image Output: ROI
- Image Output: Metadata
- Tensor Input Size: Max. 640 x 640 (H x V)
- Data Type: int8 or uint8
- Memory: 8,388,480 bytes (firmware, network weights, and working memory)
- Framerate: 2x2 binned: 2028 x 1520 10-bit @ 30fps
- Framerate: Full resolution: 4056 x 3040 10-bit @ 10fps
- Dimensions: 25 mm x 24 mm x 11.9 mm
- Cable Length: 200 mm
- Cable Connector: 15 x 1 mm FPC or 22 x 0.5 mm FPC
- Operating Temperature: 0°C to 50°C
A strong choice for makers and developers building Raspberry Pi vision AI projects, from smart cameras to robotics and monitoring systems.
Jargon buster
Plain-language definitions for the technical terms used above.
- FPC
- FPC stands for flexible printed circuit, a flat flexible cable or connector style often used where space is tight. It matters because this breakout needs the correct pin count and pitch FPC cable to connect reliably to the display or high-speed interface.
- ISP
- An image signal processor is hardware that helps process raw camera data into usable images or video. It matters for vision projects because it can improve camera handling and reduce the processing load on the main CPU.
- RGB
- Short for red, green and blue, usually referring to an LED that can mix those three colours. It matters because controlling an RGB LED teaches how separate outputs combine to create different colours.
Supplier page — dfrobot.com
Supplier Description · 160.2 KB · Click any page to view full size
